The HEKAT app

Our story We are team of engineers, physicists and doctors from Italy, under the wise guidance of a legal consultant and the creative mind of an app developer. Some of us love biomedical research and are committed to it working in research hospitals and universities in Italy and in other EU countries. What we all have in common is the uttermost desire to share our expertise and skills with EU countries (and beyond!) and to give our contribution for fighting COVID-19.


Coronavirus disease 2019 (COVID-19) is an ongoing virologic infection caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). So far, more than 2.74 million cases of COVID-19 have been reported and it is currently spreading in more than 180 countries.

A recent work by the Imperial College London estimated that, across 11 European countries, more than the 11% of population is affected by COVID-19.


Given the abrupt explosion of the pandemic, the SARS-CoV-2 is still under study, and only few symptoms have been officially associated to the virus, including fever, dry cough and shortness of breath. Nevertheless, many infected people may be asymptomatic, i.e., infected without displaying evident symptoms. Asymptomatic patients represent the hardest detectable source of transmission.

Can we track the asymptomatics down? Of course we can, but we need your help!


We want to provide you with a safe app, endowed with state of the art privacy protocols, supported by a decentralized server, and bluetooth trackers - but only after getting your consent! Help us train our model predicting your safety net with the highest accuracy!

The app we propose allows assigning each people a percentage value linked to their probability of having contracted the virus. Specifically, you will be requested to insert your health status, that will be sent anonymously to a server. In this way, you will contribute to train an adaptive mathematical model which predicts the probability of being infected. You can insert your data manually or make our app talk with other E-health devices.

Moreover, through a Bluetooth Low Energy (BLE) approach, a device can also read the probability of the other devices it connects to. This adjunctive data can be used as further input for the mathematical model, leading our outcome to be as accurate as possible. Notably, this information remains allocated on user smartphone, without being transmitted to the server.

Finally, a geographical map showing the distribution of the virus probability is a further outcome of our app, that can help making population statistic and prevent new outbreaks.

Some hardcore statistics for our nerd readers

The more data will be collected, the lowest the model uncertainty. So, we plan to set up a first burn-in period during which the beta testers just collect data to create the first shape of the model. Once collected their data, a survival analysis will be performed, e.g., by developing a Cox-proportional hazard model, that will allow to determine which are the most significant predictors.

Once selected the main features, they are given as input to a neural network, to take into account nonlinear combination of them too. The model estimates vary dynamically, whenever a new data is read by the app, in order to converge to the actual population probabilities.

The probabilities estimated by this model can be then used as transition probabilities of an epidemiological model, e.g., the Susceptible-Exposed-Infectious-Recovered (SEIR). This will allow improving the knowledge about the virus spreading process, limiting the disease dispersion.


Our project borns for facing with coronavirus necessities. Hopefully,vaccines will be developed soon and the number of infections will reduce. Which are the cities that need a huge number of vaccines? It can be answered by using the geographical map provided by our app. Moreove, since so far, information on the duration of immunity status is uncertain, our app can help having a map showing the distribution of vaccinated people who are still immune. Nevertheless, we hope the pandemic will end soon (especially if using our proposal). Then, the novelty strenghts that our app brings on the maket can be adapted for many other purposes, linked to building new specific networks, based on users passions and hobbies.


We would never have believed that three days was enough time to do all this. We are not referring only to the final product, but to the feelings we shared with each other. Collaborating, focusing, discussing and learning are difficult tasks to do when working all together in the same place. We were able to do it remotely, by sharing our "quarantene lonliness", motivated to help the world fighting and rising again.

We would to strongly thank our mentors and skill mentors for their effort: Venkatraman Ravi, Antonio Pelliccia, Jennifer Relic.

Share this project: